Parallel implementation of QRD algorithms
145.
B. B. Zhou and R. P. Brent,
Parallel implementation of QRD algorithms on the Fujitsu AP1000,
Proc. 17th Australian Computer Science Conference
(edited by Gopal Gupta),
Australian Computer Science Communications 16 (1994), 733-743.
Also Technical Report TR-CS-93-12, CSL, ANU,
Nov. 1993, 11 pp.
Abstract:
dvi (3K),
pdf (67K),
ps (28K).
Paper:
dvi (22K),
pdf (166K),
ps (73K).
Abstract
This paper addresses several important aspects of parallel implementation of
QR decomposition of a matrix on a distributed memory MIMD machine, the Fujitsu
AP1000. They include: Among various QR decomposition algorithms, which one
is most suitable for implementation on the AP1000? With the
total number of cells
given, what is the best aspect ratio of the array to achieve optimal
performance? How efficient is the AP1000 in computing
the QR decomposition of a matrix?
To help answer these questions we have implemented various orthogonal
factorisation
algorithms on a 128-cell AP1000 located at the Australian National University.
After extensive experiments some interesting results
have been obtained and are presented in the paper.
Comments
For related work see
[136,
137,
138].
Go to next publication
Return to Richard Brent's index page